# Visual Understanding on JARVIS-VLA Benchmark 1.0 (test) Source: [Visual Understanding on JARVIS-VLA Benchmark 1.0 (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:27:05.057Z Evaluation of visual understanding capabilities using the test split of the JARVIS-VLA Benchmark 1.0. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Visual Understanding | | Dataset (including any stated split) | JARVIS-VLA Benchmark 1.0 (test) | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 10 of 10 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [Accuracy](): Higher is better; unit: PERCENT. - [Rank](): Lower is better; unit: SCALAR. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | GPT-4o | 76.7 | Not recorded | 68fa7b3e1e209131cea8747c | | JARVIS-VLA-Qwen2-VL | 76.7 | Not recorded | 68fa7b3e1e209131cea8747c | | Qwen2-VL (Vision) | 65.1 | Not recorded | 68fa7b3e1e209131cea8747c | | GPT-4o-mini | 62.8 | Not recorded | 68fa7b3e1e209131cea8747c | | Molmo-d-0924 | 58.1 | Not recorded | 68fa7b3e1e209131cea8747c | | Qwen2-VL (Grounding) | 51.2 | Not recorded | 68fa7b3e1e209131cea8747c | | Qwen2-VL | 46.5 | Not recorded | 68fa7b3e1e209131cea8747c | | Qwen2-VL (Knowledge) | 46.5 | Not recorded | 68fa7b3e1e209131cea8747c | | Llama-3.2 | 44.2 | Not recorded | 68fa7b3e1e209131cea8747c | | Llava-Next | 41.9 | Not recorded | 68fa7b3e1e209131cea8747c | ### GPT-4o Accuracy: 76.7. Result ID: 698b83687a967eddb76f3973. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### JARVIS-VLA-Qwen2-VL Accuracy: 76.7. Result ID: 698b83687a967eddb76f397c. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Size":"7B"}. This record does not assert independent reproduction. ### Qwen2-VL (Vision) Accuracy: 65.1. Result ID: 698b83687a967eddb76f397a. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Size":"7B"}. This record does not assert independent reproduction. ### GPT-4o-mini Accuracy: 62.8. Result ID: 698b83687a967eddb76f3974. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Molmo-d-0924 Accuracy: 58.1. Result ID: 698b83687a967eddb76f3976. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Size":"7B"}. This record does not assert independent reproduction. ### Qwen2-VL (Grounding) Accuracy: 51.2. Result ID: 698b83687a967eddb76f397b. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Size":"7B"}. This record does not assert independent reproduction. ### Qwen2-VL Accuracy: 46.5. Result ID: 698b83687a967eddb76f3978. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Size":"7B"}. This record does not assert independent reproduction. ### Qwen2-VL (Knowledge) Accuracy: 46.5. Result ID: 698b83687a967eddb76f3979. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Size":"7B"}. This record does not assert independent reproduction. ### Llama-3.2 Accuracy: 44.2. Result ID: 698b83687a967eddb76f3977. [Reporting paper]() Recorded evaluation configuration: {"Model Size":"11B"}. This record does not assert independent reproduction. ### Llava-Next Accuracy: 41.9. Result ID: 698b83687a967eddb76f3975. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Size":"8B"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.